TheIstar
-
Posts
29 -
Joined
-
Last visited
-
Days Won
1
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by TheIstar
-
-
9 hours ago, kizer said:
Lets say you want all your files to move when your drive is 50%. So the Plugin will move your files when it is 50%
Lets say you want all of your files to move when your drive is 50%, but at the end of the month before Parity is ran you want all your files in the array regardless of how full your SSD is. You have that option by using the Move on a Schedule within the Plugin, but you can't move on a schedule with the original unraid mover because the Plugin supersedes it.
in all sense the old Mover is not used anymore unless you click the Move button within the plugin settings in the bottom right corner which will trigger the default unraid mover.
Yes, it would be great if it worked like that, but as other people have also already pointed out, the move on a schedule does not seem to move all files anymore unless the cache is filled to the percentage you set. That is why I asked what the use of it is now.
-
Maybe I am not understanding the option right, but what is the added value of the "Force move of files on a schedule" if it does the exact same thing as the mover schedule under the normal mover settings.
They way I had this setup before was that the normal mover settings checked every hour if the cache was over a certain threshold and if it was it would move files off. That way the cache never became totally full and this was ideal to prevent issues with vm's, dockers or app data. But it would only do this if it was really needed.
And then in the night I had the force move all files running so it moved all files to the array when the server was not in use and was not bothering anyone.
They way the plugin is setup now there seems now way for me to keep this functionality?
Or am I missing something here or misunderstanding something? Is there a reason this is no longer possible? Would it not be possible to have the force move option invoke the old mover?
I am asking to see how to best use this new plugin functionality as currently I might just not understand it.
-
On 3/28/2021 at 8:17 PM, hugenbdd said:
I haven't tested this. It's left over from the original script. My best guess before looking at the code is that it calls the new mover, so would follow the config settings. (i.e. cache %)
I will review this piece of code and see if needs to be updated. If you must move all files, I would suggest using the User Scripts plug-in and referencing the "mover.old" file. (I think this was discussed a few weeks back).
Any change of getting is functionality back? As it is now the force move option does not do much and I really prefer to move the files of my cache in the night instead of during the day when it fills up and the array is in use.
-
Aah yes I found it.
So if I would set those myself when first running the container would that overwrite those in the file, or would the file overwrite whatever I set in the run command?
-
Hi Saarg,
I never said it was not supported, I only pointed out that the developer warns against using it. I think it is normal then for me to worry that in some months I might have an issue where my data is stuck in a not supported database. It's called planning ahead.
And again, I was not criticizing anyone for it I was only asking why this decision was made and if it was deliberate. And if there was a way for me to setup a different DB if I want to.
I did see the config file, however I am/was worried it might be problematic to edit this as I can only edit this after first starting the container. Not sure if changing the DB at that point might break something.
Do you know what env vars are missing?
-
Question regarding usage of this container.
It seems this container comes with a SQlight DB already set up. However the documentation of Wiki.js actually advised not to use SQlight and use PostegrSQL instead. In fact it mentions:
QuoteThese engines (MySQL, MariaDB, MS SQL Server and SQLite) may not be supported in the next major version of Wiki.js. Make sure you understand the implications of migrating your database to PostgreSQL if you plan on upgrading to 3.x+ in the coming years.
SQLite is not recommended for production deployments.
So the question is, does this container support the use of other DB back-ends? And if yes, does it use the environment variables mentioned in the official documentation or does it work differently?
Another reason to use PostegreSQL would be the search function as the one included with SQlite is much more limited compared to PostegreSQL.
As normally @linuxserver.io knows what it is doing I would be curious to know why the choice for SQLight was made and the use of another DB not documented. Was this a deliberate choice or maybe an oversight?
Not criticism, just genuine curiosity as I would like to understand the container and it's use case better. -
9 hours ago, Selmak said:
Late to the party.
I love unraid I have been using it for years, I have even recommend it to a family member and a friend who have gotten licenses.
I have even contributed in my own little way.
I have read your analogy twice and all I am getting from it is you think @CHBMB is acting like a Spoiled brat and they should only expect thanks from the community. Even know they and other like them stepped up to help the community (parented) when the community were being ignored from the Unraid team (parents).
Now that the unraid team has decided to parent again the uncle should expect no thanks and if other community members don't like it they can leave?
Of course their feelings are hurt they have put in so much time helping people just for the sake of helping the community.
Just like you said why should they help the kids if the parents aren't grateful and are scared their help will damage the kids?.
Remember they aren't getting paid like the unraid team.They don't have to help and share their toys.
With that being said maybe unraid will put more of an effort in when it comes to the community devs they are after all what make unraid what it is imo.
Hi Selmek,
I think you missed the important part in my analogy, namely the "ask yourself."I meant to make people think for themselves and not just follow what others, including me, think. And it seemed at the time everybody was bashing @limetech and I wanted to make people stop and think if this was fair or if the truth was more moderate.
As for my analogy and the thankfulness of the parent. Of course I think he (the developers) should be thankful.
But I would never take away a present I gave to my nephews or nieces just because I had a fight with my bother or sister. And if I give them (my nieces or nephews) a present I do not expect thanks from my brother or sister either, though it would be nice of course.
In the end it think it comes down to entitlement en expectations.Experience has learned my something though: When you do something that no one asked you to do and you do it out of your own accord and free will, do not expect any thanks, you might be thoroughly disappointment. And if you do do something because someone asked, only expect thanks from the person who asked and not anybody else. After all, that someone else might not even be happy you did it in the first place.
It seems though some people here do expect that someone else to say thanks, and there is the problem, because expectations create disappointment.
As for me thinking @CHBMB is a spoiled brat, I don't think that at all. I only think he should evaluate who should be thankful and if what he did was fair to the people he wanted to help in the first place. The people who SHOULD be thankful are all of them screaming at him to keep updating his build.
And I do actually get his frustration, I also have been frustrated in the past when people have not seen the effort I did for something. You feel sad and disappointed. I think everyone with a certain age and work experience knows. Hence me previous remark regarding expectations and disappointment.
Something I do not get from your text thought, first you say: "...they have put in so much time helping people..." -> have to.Later you say: "They don't have to help and share their toys." -> Don't have to.
Which is it now, have or don't have? if they don't then neither Limetech nor the community should have expectations of them and vice versa. If they have to, then we can expect them not to take away the toys and behave more like our employees.
The answer of course is they don't have to. It is their choice. But if someone chooses do to something it is their choice. you can not force someone else to follow it. That's the great thing about open source, you can choose to use it, you can choose to make plugins for it, you can choose to share the plugins and you can choose to stop sharing them. But you can not force someone to do anything including being thankful.I see this a lot in the opensource communities: I made this, I spend a lot of time on this and now you have to include it and you have to be thankful. The thing is, things don't work like this and it is a toxic way of thinking. And UnRaid is not even a fully opensource project.
I love UnRaid and I love all the community developed plugins and I am super thankful to all of them. I even think they deserve part of the money you pay for the product since they have such a big part in Unraid's success (idea for a community app, a community donate function?) But I have no expectations of them to keep doing it. If Binhex or LSO decided tomorrow to stop then so be it. It would suck but it is their right. So why do they have the right to expect anything, even a thanks? It is a 2 way street. Like i said, I am thankful but I should not be forced to be thankful. I am because I choose to.
I do agree with 1 thing though. It would be great if the developers put in effort to work closely with the community as objectively it is something that contributes greatly to the success of UnRaid. I just don't see it as something they have to do, but as a nice to have.
Though it seems they are already working on it and I am not in a position to judge if it is sufficient. But I am hopeful.-
1
-
-
4 hours ago, limetech said:
We do have a special forum board called "The Lab" which is available to Community Dev's and Moderators where we can and do talk about development, though we (Lime Tech, Inc) need to do a better job at posting roadmaps in there (working on it). There is another board called "The Bar" where Moderators can hang out and discuss forum issues. Finally @SpencerJ servers role of "Community Manager", though in this particular case I thought it best to address the Community directly myself.
Then I stand corrected. And thank you for doing so.
-
I've have been following this and the other thread with very mixed feelings and I feel the community is unjustly hard towards @limetech. Sure some things could have been handled better, yet I keep the feelings that the bigger injustice is not actually committed by him.
In order to understand things better and to see things from a different perspective I personally like to make analogies. Sometimes it gives different insights into situations. And I cam up with the following for this one:
We have 3 parties here, The parent (@limetech), the uncle (@CHBMB and the like) and the kid (the community).
Now the situation is that the kid is asking the parent for this shiny new toy, but for whatever reason the parent is not buying the kid the toy. Maybe it is to expensive, maybe he is waiting for the birthday, whatever..
However, the uncle who hears the kid decided to get the kid this new toy, because he loved the kid and wants to please the kid.
Fast forward and the parents sees that the kid really loved the toy but unfortunately the toy has some sharp edges and the parent is afraid the kid might hurt himself hence the parent decided to order a better and safer version of the toy.
However, when the parent tells the kid it ordered this new toy the uncle hears the parent and flies into a rage because the parent did not tell the uncle that he/she was going to buy the new toy and the uncle thinks the parents is ungrateful because he/she did not even thank the uncle.
In his rage therefore the uncle takes the toy away from the kid even before the new toy arrived (it is after all still in beta). Not only that but takes away the other toys he got the kid as well and says he is never going to give the kid any more toys. All this to punish the parent.
Now with this analogy, ask yourself.
Is the reaction of @CHBMB (the uncle) proportionate and justified? Does a parent (@limetech) need to inform the uncle of these kind of things? Sure it is nice, but is it really needed? Do you think it is right for the uncle to punish the kid? Should the parent even be grateful that the uncle presents the kid a toy with sharp edges (I know I wouldn't).
The only one the uncle should expect thanks from i.m.o is the kid. The community is and was grateful. Yet @CHBMB is the one who decided to punish the community and take away their toy because of his hurt feelings. Yet the only one who gets shit is @limetech. If I where him I would be more than a little pissed and disappointment and I think it shows in his messages.
Please read my analogy again and ask yourself who in the story did anything to hurt the kid? The parent or the uncle?
And please also think about the fact that we have no way of knowing if @limetech was not going to thanks @CHBMB for the work in an official release note, which this wasn't.
Now I do think the parent should have said something to the uncle. And I also am a bit disappointment to learn that even though UnRaid builds heavy on the community there is no special channel in place to facilitate communication with reliable community develops. Considering how well the development of both UnRaid and the community add-ons go together I kind of assumes something was already in place. However it seems this is something that is considered and worked on now.
But in everything that happened, this simple miscommunication seems far the lesser evil here. And I do think it might be good that the community asks itself again who really is to blame for taking away it's shining toy with sharp edges and if it is reasonable to have this reaction.
But that's just my 2 cents.
-
18
-
5
-
-
43 minutes ago, MothyTim said:
Well yes I'm using reverse proxy, I thought it must be a temp file and have said so before but no one suggested that! Which config file and I assume I need to passthrough a path to a folder or can I use unraids temp folder?
Many thanks
Tim
I just added the following line to my letsencrypt nginx proxy config:
proxy_temp_path /config/proxytemp 1 2
This way it just creates a sub folder called proxytemp in the letsencrypt folder in appdata.
-
On 6/12/2020 at 9:22 PM, dtctechs said:
I finally got the app to update and remove 'not available' tag. I've tried several different ports including 6595/WebUI with no luck. Whenever I attempt to hit the UI it times out.
Regardless of UI port I get this in my log.
Did you change it here? :
-
On 6/13/2020 at 10:48 AM, Goldfire said:
I have this same problem as well, I'm using external storages that point to my /mnt/ locations. I had 10GB for my docker.img file which was about 50% full, someone downloaded a file from me that came in at 4.5GB - it basically crippled and trashed my docker.img to the point where I had to delete it and recreate it from scratch. Painless process with the auto template usage from the apps tab, but it's still a pain when I have to take all the dockers down just to do it.
The only thing I can think of in the short term is to increase the docker.img size to something larger than your single largest file and inform your users to not download multiple files as an archive.
@MothyTim What are you using to replace Nextcloud? I found Nextcloud to "tick" all of my boxes but this problem is just a big strike through them all.
Since it is only when downloading/uploading it might just be the temp folder.
Did you set proxy_temp_path in your nginx config in your reverse proxy? Are you even using a reverse proxy? -
9 hours ago, jserio said:
Hey guys. The latest Deemix docker seems to be broken. I noticed the port in the config still says 9666 but the documentation mentions 6595. Regardless of what I put in the configuration, I am unable to connect. Removing and adding the container doesn't seem to fix it.
In the template add a new port 6595. You will also need to change the port for the webui (visible when you enable advanced).
If it still does not work try a force update. remember it will take some time to install the software again (you can check to log to see the progress)/.
-
15 hours ago, DontWorryScro said:
I updated to the new repository the other day but started noticing that it was showing up as "not available" in the status column of the docker list. It still launches the webui fine and i can run searches and complete downloads. I'm just not sure why it's not available. I imagine this means I am no longer able to get any sort of update?
Same for me. Did you get this sorted?
-
Just downloaded and installed the Deemix docker and it looks really cool.
Just having 1 issues, when music is downloaded and I want to change something or remove something by accessing the share it seems I do not have the rights to do it.
Looks like only the original user (nobody) has the rights to change or delete the files.
Would it be possible to change it so all users can modify/delete files?
-
On 5/10/2020 at 8:06 PM, Josh.5 said:
If you are willing to test it, I can buil a dev- tag docker image for you to download.
I had issues on my laptop getting it playing nicely.Any news on this?
-
Sure, I can test.
What kind of issues where you facing?
-
Any word on this working with quicksync/QSV now it works with nvenc? The template seems to suggest it already does...
-
13 hours ago, knex666 said:
It was not possible to set this script to post execution. The container will not start up then.
So please if you update your container run the docker exec. To update your container please change the repository.
So now you can change to nextcloud:18.0.2-apache and after pulling execute docker exec. I dont know how to make it more confortable.
Sorry
I found a way to get it running in Unraid !!!!
In advanced view under extra parameters put: --user 99:100 --sysctl net.ipv4.ip_unprivileged_port_start=0
This makes sure you run the container with the right UID and GID
Then under Post Arguments put: && docker exec -u 0 Nextcloud /bin/sh -c 'echo "umask 000" >> /etc/apache2/envvars'
This adds the right umask to the envvars file.
Now new files created have the right access rights for Unraid:
Please let me know what you think?
-
1
-
1
-
-
Does this mean I have to run this exec manually after each pull myself? Or will you automate this somehow? Not sure I understand correctly.
-
20 hours ago, knex666 said:
Ouch, ok.
Got an idea. I think we can add this changes to the Container by mounting a Script to the Container and Post execute it if the Container spawns.
Can you provide a Script that does these changes?
Cheers
No scripting expert, but I guess something like this would work:
grep -q 'umask 000' '/etc/apache2/envvars' || echo "umask 000" >> /etc/apache2/envvars
I tested it and it seems to run as expected. Also adding this line gives the required rights to everybody. I could edit and change files created by Nextcloud without issues.
-
44 minutes ago, knex666 said:
Or you can add the docker user to the users group. That would be a very straight forward solution.
I already set the docker user to 99:100.
The problem is that the umask of netxcloud user is set to 022. Which means only the user has write permissions and not the group. It should be 000 for Unraid.As I said in my previous post, you can do this by editing a file inside of the docker. But Any update to the docket would overwrite this.
-
On 3/11/2020 at 9:37 PM, TheIstar said:
Thanks for the reply,
I did already try to add these to the template. however this did not work. it seems the container is not setup to work with the variables?
some googling let me to add this to the extra parameters: --user 99:100 --sysctl net.ipv4.ip_unprivileged_port_start=0
However, even though now the correct user is running with the correct group, it looks like it does not have the correct umask set.
you can change it by bashing into the container and changing it in /etc/apache2/envvars, however, i am afraid any update to the container would overwrite it again.
Also it seems a bit overly complicated.. there must be an easier way to do it.
I did also try to install the smb client in the container, as it seems this is not included. Adding the shares as an SMB share does grant the correct rights but of course, also here any update would remove the client again. It seems there is a way to add it (see: https://github.com/nextcloud/docker/tree/master/.examples ). However I have no idea how to get this working in unraid.
Further investigation shows the Linux Server Docker actually has the same behavior if you add the unraid shares as local Storage..
As they respect the UID and GID it seems there is more to it then just adding this to the docker.
I think getting SMB to work in the docker would probably be the best/easiest solution. But I have not figured out how to do this yet...
So, any help or insight here is appreciate. -
Thanks for the reply,
I did already try to add these to the template. however this did not work. it seems the container is not setup to work with the variables?
some googling let me to add this to the extra parameters: --user 99:100 --sysctl net.ipv4.ip_unprivileged_port_start=0
However, even though now the correct user is running with the correct group, it looks like it does not have the correct umask set.
you can change it by bashing into the container and changing it in /etc/apache2/envvars, however, i am afraid any update to the container would overwrite it again.
Also it seems a bit overly complicated.. there must be an easier way to do it.
I did also try to install the smb client in the container, as it seems this is not included. Adding the shares as an SMB share does grant the correct rights but of course, also here any update would remove the client again. It seems there is a way to add it (see: https://github.com/nextcloud/docker/tree/master/.examples ). However I have no idea how to get this working in unraid.
-
1
-
[Plugin] Mover Tuning
in Plugin Support
Posted
This is awesome! Thanks for that.